On 23 June 2010, the Food and Drug Administration (FDA) and Centers for Medicare and Medicaid Services (CMS), the federal agency issuing national coverage determinations (NCDs), signed a memorandum of understanding (MOU). The proposed joint venture has caused quite a stir within the medical device sector, with many US industry officials remaining sceptical of these first steps towards a combined review for marketing approval and Medicare coverage.
With arguably different core values on data sharing, the two largest regulatory bodies in medicine appear to be unlikely collaborators, and the resulting 'parallel review' process has received mixed feedback. This is despite its aims to bring innovative medical devices to market more quickly and improve patient access in response to complaints about the separate, lengthy frameworks currently in place to achieve FDA and CMS clearance.
Currently, US manufacturers must approach both of these agencies independently for approval, a palpable reason why the sector is seeking a speedier and less complicated model. The new process is quicker as it promotes data sharing between the FDA and CMS, reducing the time it takes to receive coverage and payment for some devices.
An MOU is already in place, meaning that the agencies have agreed to work together, thus strengthening their ability to achieve their respective missions. So what exactly will a 'parallel review' process permit? Although it sounds intriguing, much of the industry does not consider the proposal necessary and fails to see which problems it will actually solve.
James Mazzo, vice-president of the Advanced Medical Technology Association (AdvaMed), has recently said that although it is still reviewing the details and believes that the collaboration is "well-intentioned", the agencies could be agreeing to more than they can manage.
The Medical Device Manufacturers Association (MDMA) has taken a similar stance. It recently expressed concerns over the scheme via a letter sent out to Donald Berwick, administrator of CMS, and Margaret Hamburg, commissioner of the FDA.
However, Thomas Novelli, vice-president of government relations at MDMA, insists that the group's uncertainty has not impacted on its close relationship with either party. "We continue to work with both companies and try to provide useful advice and guidance," he says. "We're going to continue to work with them, if they are willing to listen."
While the MDMA considers the 'parallel review' system to be flawed, it does commend its attempts to speed up the route to approval. "The FDA process can be anywhere from four to five years depending on the type of product, and given the uncertainty of the FDA right now it can be even longer than that," Novelli says. "Following this, medical device manufacturers look to get their product reimbursed from one of the federal healthcare programmes and the process for NCD can be anywhere from two to three years. We agree that the entire process needs to be expedited."
Indeed, the prospect of securing reimbursement more quickly is news that the whole industry can get on board with.
"Bringing a medical device to market is a very costly process, ranging anywhere from $50m to $100m," Novelli notes. "A shortened timeframe would hopefully be a more efficient use of precious resources. Reducing the time to three or four years is something everyone would be in agreement with."
Why is the industry so nervous about FDA and CMS collaboration? The new relationship formed between the agencies is complicated due to its different treatment of proprietary information. The sector is concerned about data held by the former seeping out if it is shared with CMS, since it does not have the same statutory obligations for confidentiality and is subject to transparency requirements. In addition, innovation and investment support will occur only if the confidentiality of manufacturers' information is maintained throughout the development process.
"Greater assurance that the information in these applications is protected is crucial," says Novelli. "An NCD process opens a company up to national review, which in most cases means the information becomes public and is therefore available to competitors. CMS would need to make amendments so that manufacturers feel comfortable about participating in 'parallel review'. There doesn't seem to be a strong mechanism in place to protect that information in the proposal."
While it is expected that collaboration will enhance the ability of each agency to carry out their responsibilities, industry officials are worried that the partnership could have other side effects. One such concern is that the process will not ensure that companies adhere to strict conditions defined in law and regulation.
In addition, many companies do not trust the agencies to play fair with shared information; they believe that regulatory decisions and cost could influence whether or not a device makes it through the approval process.
A further reason why NCDs are scrutinised to such an extent is the inflexibility they present for the device industry. The MDMA, for instance, has emphasised that this should not be the only pathway to coverage, and Novelli says that the organisation is "not happy with the NCD review process in place".
"With NCDs it's one bite at the apple," Novelli continues. "If you don't get it, then that's it. It's really risky, especially if you're a small innovator. If you get approval at an NCD level then Medicare will reimburse countrywide. The reason why it's a big gamble is because if you don't get that then the LCD (local coverage determination) will not be approved either, because there is a national non-coverage resolve."
In addition, the current plans implementing NCDs could result in significant burdens for manufacturers, and have a harmful effect on innovation. Smaller medical device companies, often responsible for the development of breakthrough medical innovations, are the ones likely to fall short. For example, timing of an NCD is critical because a review initiated before the manufacturer has built a sufficient body of evidence for CMS review could result in a non-coverage determination denying beneficiaries access to the technology and discouraging further investment in research on that equipment. The MDMA has also stated in its letter that the proposal does little to improve access to the innovative technologies developed by these companies because LCD models already work sufficiently.
Notably, many manufacturers are cautious about pursuing coverage through an NCD and turn to other options. In fact, NCDs are merely one of the three ways coverage of a device or service is agreed upon. "Many medical device companies choose to go through an LCD route," Novelli says. "The manufacturers go to different regions of the county – for example, in the south-east there is an LCD process and if the product is approved at that level then states such as Florida, Georgia and Alabama would be able to receive reimbursements through the process."
Coverage issues are not the only obstacles for manufacturers looking to bring new devices into the market. The MDMA also asks the FDA and CMS to address various issues surrounding coding in order to minimise significant delays between the regulatory and reimbursement process.
Novelli says that the organisations are defeating the goals they are attempting to achieve by not turning their attention to these potential hurdles. "Many companies are finding it increasingly difficult to get healthcare common procedure coding system (HCPCS) codes and current procedural terminology (CPT) codes for their technology," he says. "If the agencies are looking to increase the process then the proposal should address the issues concerning coding, which it doesn't."
Possible opportunities for third parties to initiate a request for the process are a further concern. The MDMA has asked that it be voluntary and that manufacturers be allowed to withdraw their application at any time. Novelli says that, in addition to confidentiality issues, allowing outside parties to initiate the process could also slow the advancement of a manufacturer's technologies.
"The problem we see, and this goes back to the issue of NCD, would be that if you are a small company with an innovative product there is nothing to prevent a larger medical device company or competitor instigating the review for a specific technology," he notes. "We're just waiting to see where they go with this. Generally, we believe the concept makes sense, but the agencies need do it intelligently and effectively, while addressing the big issues."
An unclear future
Nothing is set in stone just yet, however. FDA-CMS data sharing and 'parallel reviews' have been discussed for years and the US government has previously announced near-term plans, with no result. The idea was first proposed after Mark McClellan left his post as commissioner of the FDA, becoming an administrator of CMS back in 2004, but the plans eventually came to nothing.
While it is unclear whether or not the collaboration will materialise, it's evident that holes surrounding confidentiality, coding and innovation need to be satisfactorily filled in order to keep the medical device industry onside.
Final thoughts: Parallel lines of enquiry
In his letter to the FDA and CMS in December 2010, Thomas Novelli provided the following comments regarding the proposed 'parallel review'.
- A parallel review process that is focused on development of NCDs may not improve access to many innovative devices.
- Any parallel review process must be voluntary for the manufacturer and must not be initiated by a third party.
- The process must also protect the confidentiality of manufacturers' proprietary information.
- The process must not delay FDA approval or clearance of new technologies.
- The process should address coding and reimbursement in addition to, or instead of, national coverage.